Throughout the COVID-19 pandemic, people experiencing homelessness have faced a unique set of challenges in protecting themselves and their families from the virus. People experiencing homelessness are already at higher risk for COVID-19 due to underlying health risks and other factors, while homelessness itself creates barriers to meeting social distancing guidelines and accessing testing and treatment. As with other congregate settings, shelters are at high risk for COVID-19 spread. Individuals who are unsheltered face additional risks from the cold weather and lack of access to hygiene and sanitation facilities. This brief explores issues related to risks related to COVID-19, vaccine priority in state plans, and other policy options that affect access to vaccines for people experiencing homelessness.

COVID-19 Risks for People Experiencing Homelessness

As of January 2020 (the latest comprehensive data available, but prior to the pandemic), approximately 580,000 people in the U.S. were experiencing homelessness – about 18 out of every 10,000 people on a given night (Appendix Table 1). The majority are individuals, but 30% are people in families. Nearly one in five (21%) are chronically homeless. The U.S. Department of Housing and Urban Development (HUD) considers an individual to be homeless if he or she lives in an emergency shelter, transitional housing program (including safe havens), or a place not meant for human habitation, such as a car, abandoned building, or on the streets. According to point-in-time data on homelessness from HUD, homelessness increased by 2.2% in 2020, marking the fourth year of national increases after long-term downward trends in homelessness since 2007. Data show there were higher rates of homelessness among people of color, specifically among Black or African American people, American Indian and Alaska Native people, Native Hawaiian and Pacific Islander people, and Hispanic or Latino people. For example, African Americans account for approximately 12 percent of the general population but 39 percent of people experiencing homelessness. Fifty-four percent of people experiencing homelessness are in four states (California, New York, Florida, and Texas) and over a quarter (28%) are in California alone.

Because many people who are homeless have underlying medical conditions, they may also be at higher risk for severe illness from COVID-19. Data and research show that people who are homeless have a range of health conditions ranging from physical and mental health problems to substance use conditions and tri-morbidity (co-occurring physical health, mental health, and substance use disorder challenges). These challenges present at higher rates for the unsheltered homeless populations. In particular, people experiencing homelessness are more likely to suffer from diabetes, heart disease, and HIV compared to the general population. People with these conditions face (or may face) a greater risk of developing severe illness from COVID-19 or of dying from COVID-19.

Homelessness services are often provided in congregate settings that can lead to rapid spread of infection. As of January 2020, more than six in ten (61% or 354,386) people experiencing homelessness were sheltered, meaning they used an emergency shelter, safe haven, or transitional housing program.   While similar data for 2020 are not available, data show that in 2018, 1,446,000 people experienced sheltered homelessness at some time during the year, four times the number that were counted as sheltered homeless in January 2018. Early in the pandemic, a Centers for Disease Control and Prevention (CDC) study of residents and staff members from five homeless shelters in Massachusetts, California, and Washington found that high proportions of residents and staff members had positive test results for COVID-19. Updated guidance from CDC on responding to COVID-19 for homeless populations calls for communication and coordination among community partners and identifies strategies to reduce transmission at homeless shelters, such as reconfiguring the layout to maintain social distancing among residents and staff, identifying overflow sites to reduce crowding, developing policies and locations for isolating people with COVID-19, ensuring staff who have contact with residents with COVID-19 wear personal protective equipment (PPE), and improving ventilation systems.

Unsheltered people experiencing homelessness are also at high risk from COVID-19. As of January 2020, approximately 226,000 people (or 39% of the total homeless population) were unsheltered, meaning these individuals sleep outside and in other locations not meant for human habitation. According to the CDC, while outdoor settings may allow people to increase physical distance between themselves and others, sleeping outdoors often does not provide protection from the environment, adequate access to hygiene and sanitation facilities, or connection to services and health care. CDC provides specific guidance to respond to COVID-19 for unsheltered homeless populations that includes strategies for training outreach staff on how to help prevent people from becoming sick and on how to link those with symptoms to medical care. The guidance also includes considerations for unsheltered homeless in encampments such as allowing for space between sleeping quarters or tents and working to improve sanitation.

National data related to testing, cases and deaths for those experiencing homelessness are not available. One website has compiled overall deaths from people experiencing homelessness and has added available data for 18 cities or counties reporting data on COVID-19 deaths among people experiencing homelessness. While the data are not official counts and are likely undercounts of actual deaths due to COVID-19 among this population, they provide some information specific to this population. For example, as of the end of October, the New York City Department of Homeless Services reported that 104 homeless people had died from COVID-19, including 95 sheltered individuals. While the overall counts appear to be small, analysis of the NYC data deaths for sheltered homeless per 100,000 were 75% higher than the overall rate for New York City. In Los Angeles County, as of March 8, 2021, there were 7,015 cases and 190 confirmed deaths for people experiencing homelessness out of an estimated population of 66,000. DC reported that as of March 9, 2021, there have been 521 reported positive cases of COVID-19 among people in homeless shelters and 25 individuals in the Homeless Services System have died from COVID-19. Los Angeles, Orange County, CA and Phoenix, AZ reported a surge in deaths among people experiencing homelessness in 2020; however few deaths were directly attributed to COVID-19 in these localities. The Los Angeles report showed a 25% increase in deaths from January to July 2020 compared to the same months in the prior year, but also found that a sharp increase in drug overdoses accounted for most of the increase and COVID-19 was the fifth leading cause of death among people experiencing homelessness. Some news reports indicate that the pandemic may have significantly limited access to facilities and services that may have prevented these deaths. One advocate noted that people who are homeless do not receive autopsies and another researcher noted that housing status is not listed on most death certificates or hospital records so deaths attributable to COVID-19 among people who are homeless are likely undercounted.

Data from health centers that primarily treat patients who are homeless show rates of positive tests and vaccines similar to overall health centers. Health care for the homeless (HCH) grantees are community health centers that receive funding to treat patients who are homeless. Some HCH grantees receive funding to treat only people experiencing homelessness while other HCH grantees treat both housed and unhoused patients. Cumulative data through January on COVID-19 testing and vaccinations show that HCH clinics have a similar share of individuals who test positive for COVID-19 relative to all health centers (12% compared to 13%) and a similar share of those receiving both COVID-19 vaccine doses was lower (27.7% compared to 29.5%). Additionally, the National Health Care for the Homeless Council has partnered with CDC to collect and report data from universal testing events at shelter or encampment-based service sites during the pandemic. There were 557 total testing events that participated and submitted data as of March 2021 showing a 6.1% positivity rate for clients, presumably lower than the rate at health centers because these were universal testing events.

Vaccine Prioritization, Access and Coverage Issues for Those Experiencing Homelessness

While federal guidance on COVID-19 vaccine allocation does not explicitly include people experiencing homelessness among the priority populations, it acknowledges higher transmission rates in congregate settings. In December 2020, the CDC Advisory Committee on Immunization Practices (ACIP) recommended that states prioritize certain at-risk populations for initial vaccine allocations, including health care workers and long-term care residents in phase 1a, people ages 75 and older and frontline workers in phase 1b, and people ages 65-74 and younger adults with high-risk medical conditions in phase 1c. Other ACIP guidance notes that states may choose to include people who reside in congregate living facilities, such as correctional or detention facilities and homeless shelters, along with priority groups in phases 1b and 1c due to “their shared increase risk of disease.”

About half of states include people experiencing homelessness in their state vaccination plans. Although people in congregate settings face increased risk of contracting COVID-19, only 25 states explicitly prioritize residents in homeless in shelters for COVID-19 vaccine allocation while another six states prioritize people living in congregate settings, but do not specify homeless shelters (Figure 1). Only Massachusetts and Oregon include people in homeless shelters in their phase 1a group. The remaining states include people in homeless shelters in phase 1b or 1c. Oregon also includes people experiencing homelessness (those who are unsheltered) in phase 1b and Nevada includes people experiencing homelessness, both sheltered and unsheltered, in phase 1c. With supplies of the COVID-19 vaccine still limited, states are phasing in eligibility for the vaccine, and different groups become eligible for the vaccine at different times. As of March 22, 2021, people living in homeless shelters were eligible for the vaccine in 21 states.

Outreach and incentives can help to encourage take-up of vaccines. As noted above, how states prioritize groups for the COVID-19 vaccine will have implications for access for people experiencing homelessness. State plans have been evolving over time and more states are including individuals in homeless shelters as a priority. In addition, CDC has stated that the newly approved, 1-dose Johnson & Johnson vaccine may be desirable for people experiencing homelessness for whom scheduling a second dose would be difficult and because the vaccine is easier to transport and store. Specific outreach strategies that account for a high prevalence of mental health issues and potential mistrust of the health care system will help to ensure access to and take-up of vaccines. Advocates for people experiencing homelessness also have said incentives such as gift cards, socks, or other basics could be used to help encourage take-up of vaccines. Massachusetts received a $25 million grant from CDC for an effort to reduce barriers to vaccination in hard-hit areas; of that amount $3 million will be used to fund organizations to administer vaccines to groups “not effectively reached by other outreach efforts,” including homeless people living on the streets or in encampments.

An initiative to provide direct allocation of COVID-19 vaccine to health centers will help reach vulnerable populations, including people experiencing homelessness. As part of an effort to increase equity in vaccine distribution, the Biden administration recently launched the Health Center COVID-19 Vaccine Program, which provides designated health centers direct allocation of COVID-19 vaccines. The initial phase of the program allocated 1 million doses to 250 health centers across the country that serve a significant number of particularly vulnerable populations, including people experiencing homelessness, Of the initial 250 health centers, more than one-third (35%) are HCH grantees. With vaccine supplies increasing, an additional 700 health centers were invited to participate in the program on March 11, 2021.

A number cities are also using mobile teams to conduct outreach and administer vaccines for people experiencing homelessness. For example, in Enid, OK, Gainesville, FL, Berkeley, CA, Louisville, KY, Honolulu, HI and Sacramento, CA mobile teams with staff from health departments, fire rescue and other public health groups are going to shelters, places where people who are homeless get food and areas where unsheltered homeless individuals live to administer vaccines. In DC, the Department of Human Services is partnering with the entity that is the main provider of medical care to the homeless to administer vaccines in homeless shelters. These efforts recognize people who are experiencing homelessness are not able to sign up with a computer for a vaccine appointment and then get to that appointment.

Medicaid can provide coverage and access to care (including COVID-19 testing and treatment) for homeless populations, particularly in states that have adopted the expansion. Broader coverage exists for people experiencing homelessness in states that have adopted the ACA Medicaid expansion. Based on data from HCH programs in 2018, the overall rate of uninsured was 34%; however, in states that adopted the expansion, the rate of uninsured patients was 23% compared to 66% in non-expansion states; however, even in expansion states the rate of uninsured at HCH programs range because all those who are eligible for Medicaid may not be enrolled. In focus groups following the implementation of the ACA, providers serving people experiencing homelessness reported that the coverage gains from expanding Medicaid enabled patients to access many services that they could not obtain while uninsured, including some life-saving or life-changing surgeries or treatments. For people experiencing homelessness who are uninsured, The American Rescue Plan provides a new state option for coverage of COVID-19 treatment services, without cost-sharing. The COVID-19 uninsured testing group was created by the FFCRA and is available at state option, with 100% federal matching funds, during the PHE, the American Rescue Plan adds COVID-19 treatment services to this group.

Medicaid can also provide some services and supports to help people experiencing or at risk of experiencing homelessness in responding to COVID-19. For example, while Medicaid cannot pay directly for housing, Medicaid can pay for community transition costs to facilitate individuals transitioning from an institutional or other congregate living arrangement (such as a homeless shelter) to a community-based living arrangement. Medicaid coverage of community transition as well as respite care may be helpful in addressing temporary housing needs for individuals during the pandemic.

States can also address the health needs of people experiencing homelessness by requesting temporary disaster relief authorities in Medicaid/CHIP. As of January 2020, 44 states have received authority through an approved 1135 waiver to allow provision of services in alternative settings, such as homeless shelters or mobile units. Authority for these waivers is tied to the duration of the national emergency and the public health emergency declarations. Additionally, Rhode Island received approval from the Centers for Medicare & Medicaid Services (CMS) for a disaster relief SPA that added an emergency case management benefit for Medicaid beneficiaries experiencing homelessness.

Looking Ahead

While issues related to health care access for people experiencing homelessness are not new, the pandemic has exacerbated many challenges faced by this population and the total numbers of people experiencing homelessness has likely increased. People experiencing homelessness, particularly those receiving services in homeless shelters, are at greater risk of severe illness or death from COVID-19; however, national data about cases and deaths for this population are not known. Looking ahead, ensuring access to and take-up of the COVID-19 vaccine will be an important step in mitigating the health effects of COVID-19 for people experiencing homelessness. The approval of the 1-dose Johnson & Johnson vaccine along with distribution to clinics serving people experiencing homelessness and targeted outreach will help to ensure access to and take-up of vaccines. Finally, as the economic effects of the pandemic continue, recent polling shows that 16% of adults report they have fallen behind on their rent or mortgage and other data show that 7% of adults have no confidence that their ability to make next month’s housing payment. To help address problems of housing insecurity and homelessness, the American Rescue Plan provides $5 billion for housing vouchers for those at high risk of becoming homeless, $5 billion for homelessness assistance and supportive services and more than $20 billion in funding for low-income renters at risk of losing housing.

Key Takeaways

• One in ten women report quitting a job due to a pandemic related reason and almost half said that one of the reasons was because they felt unsafe at their workplace.
• School closures had a substantial effect on working mothers’ ability to fulfill work obligations. One in ten working mothers with children under 18 said they quit a job due to COVID and half of this group cited school closures as one of the reasons. Three out of ten working mothers said they had to take time off because school or daycare was closed.
• In addition to juggling new, increased home and work responsibilities, many women went without pay due to school closures. Almost half (47%) of working mothers said they took unpaid sick leave because their child’s school or daycare was closed. This rose to 65% among low-income mothers and 70% among those working part-time jobs.
• Family caregiving responsibilities before and after the pandemic have largely fallen on women. More than one in ten women report they were caring for a family member who needed special assistance prior to the pandemic. Over one in ten women report that they have new caregiving responsibilities as a result of the pandemic.
• Over half of mothers with school age children said that the stress and worry of the pandemic has affected their mental health, with one in five characterizing the impact as “major.” However, only 16% of mothers have sought mental health care. Given that the mental health effects are likely to persist for a long time, access to mental health care will be an important issue to watch, particularly for mothers.

Introduction

While the COVID-19 pandemic has impacted people across the US, changing the way we work, live, and access health care, gender-based disparities that existed prior to the pandemic have been magnified. Many women have been on the front lines of the COVID-19 emergency, as essential workers, mothers, and caregivers. Lack of paid leave, family caregiving responsibilities, traditional gender roles, and health concerns have placed many of the burdens of the pandemic squarely on the shoulders of women, falling particularly hard on women of color and those who are low-income. This brief provides new data from the KFF Women’s Health Survey, a nationally representative survey of 3,661 women and 1,144 men ages 18-64 (Methodology) conducted November 19, 2020 – December 17, 2020. Among several topics related to women’s health and well-being, we asked respondents about experiences during the COVID-19 pandemic. In this brief, we highlight how experiences during the COVID-19 pandemic related to work, childcare, and caregiving differ by gender and among different subpopulations of women.Nearly 1 in 10 women (8%) report quitting their job for a reason related to COVID-19. Larger shares are younger, Black or Hispanic,¹  uninsured, low-income, and have less than a Bachelor’s degree (Figure 1). For example, 17% of low-income women had to quit a job for a COVID-related reason, compared to 5% with higher incomes.

Almost half of women who have quit a job for a reason related to COVID-19 (48%) say they quit because they did not feel safe at their workplace. Three in ten women (30%) quit their job due to COVID-19 because their child’s school or daycare was closed (Figure 2). A larger share of women with Medicaid compared to women with private insurance (44% vs. 20%) and a larger share of low-income women compared to women with incomes ≥ 200% of the federal poverty level (FPL)²  (37% vs. 23%) say they had to quit their job because their child’s school or daycare was closed, leaving the most disadvantaged with a loss of income. Nearly a quarter of women say they quit a job because they live with someone at elevated risk for COVID-19 complications.

Half of mothers who quit a job because of the pandemic said one of the reasons was because their child’s school or daycare was closed. With in-person schools closed, mothers have taken on many new responsibilities, including even more childcare than usual, assisting with remote learning, and in some cases shifting to full homeschooling just to name a few. For mothers of young children who also work outside the home, they have been doing all of this while trying to maintain employment. While most mothers have been able to maintain their jobs, some with changes to their workload and schedules, it is not surprising that some have had to leave jobs as a result of school and daycare closures and the numerous new responsibilities that mothers now carry. One in ten women with young children (11%) say they quit a job due to the pandemic (Figure 3). Among this group, half (51%) said one of the reasons was because their child’s school or daycare was closed. Many mothers (42%) also said they quit because they did not feel safe at their workplace (Figure 4).

The share of women who report leaving a job is significantly higher among single mothers (17%) compared to those who are married or have partners (9%). Single mothers may face the double jeopardy of not having a partner to assist with childcare or another source of income, however. Almost one in five mothers who are not currently working (18%) report that they had to quit a job due to the pandemic, as do 7% of working mothers, suggesting that some mothers who had to quit a job during the pandemic have since found another one.

Mothers were more likely than fathers to miss work due to school and daycare closures. Many workers had to take time off due to the COVID pandemic, due to illness as well as school and daycare closures. The largest difference between women and men having to take time off work was because of school or daycare closure due to COVID-19 – this burden fell largely on women with school age children (30% vs. 20%) (Table 1). Low-income women were disproportionately affected, with almost four in ten (38%) taking time off because their child’s school or daycare was closed compared to 27% of women with incomes ≥200% FPL. Rates were similar among mothers of different racial and ethnic groups, except for Asian women, who were less likely to have taken time off because of school and daycare.

In addition to bearing the brunt of childcare and caregiving responsibilities, many women lost pay during the pandemic because they missed work due to quarantining or school closures. Over one-third of women (35%) say they took unpaid sick leave when they became ill with COVID-19 or were quarantining, and nearly half of women (46%) say they took unpaid sick leave when their child’s school or daycare was closed due to COVID-19 (Figure 5). Most employers offer some amount of paid sick leave and the federal government provided support for emergency paid sick leave to some employers in 2020, but these benefits were not available to all workers and are time limited. Many women may not have been eligible for paid sick leave or may have taken more time off than available for pay at their workplace.

Over two-thirds of working mothers who were low-income and nearly three in four working part-time jobs were not paid when they had to take time off because their child’s school or daycare was closed due to the pandemic. Lack of a national paid leave policy disproportionately affects women, who are more likely than men to have to take time off work for childcare reasons, and some women are particularly affected. Part-time and low-wage jobs are less likely to offer paid leave benefits, as reflected in our survey. Most women in these jobs went without pay when they took time off due to school and daycare closures last year (Figure 6).

Nearly half of employed women have worked outside their home during the pandemic. More than half of men (53%) and nearly half of women (46%) report that they are working in a location outside their home during the pandemic. At the outset of the pandemic, some companies shifted to telecommuting, but many workers do not have this option as some industries and jobs are not amenable to remote work. Working outside of the home during the pandemic is more common among women who are younger, have lower educational attainment, live in rural areas, and report lower incomes. For example, more than half of Hispanic women (53%) compared to 45% of White women, have worked outside their home. Conversely, half of Asian women workers (51%) say they are working from home (Figure 7). More than half of women residing in rural areas (58%) report working outside the home, compared to 44% of women in urban communities.

Family caregiving responsibilities before and after the pandemic have largely fallen on women. More than one in ten women report they were caring for a family member who needed special assistance prior to the pandemic. Over one in ten women report that they have new caregiving responsibilities as a result of the pandemic. A higher share of women (14%) than men (9%) say they were caring for a family member who needed special assistance prior to the pandemic. This gender difference in caregiving extends to new caregiving responsibilities due to reasons related to the pandemic (12% of women vs. 8% of men) (Table 2).

One in five women ages 50-64 say they were caring for a family member who needs special assistance prior to the pandemic, higher than women in their reproductive years (11%), but similar shares (12%) in both age groups report taking on new caregiving responsibilities as a result of the pandemic.

Women who are parents of a child younger than age 18 in their household are significantly more likely than women without children under 18 to have taken on new or additional responsibilities caring for a family member because of the pandemic (16% vs. 10%). The pattern is similar among men, with 15% of men with children under age 18 reporting they have taken on new caregiving due to the pandemic, compared to 5% of men without children. However, it is worth noting that some respondents, especially parents of young children, may have interpreted new and additional responsibilities caring for a family member or relative as homeschooling. Almost one in five low-income women (19%) were caregivers before the pandemic compared to 13% of higher-income women. Low-income women (15%) are also more likely than higher-income women (11%) to have gained new caregiving responsibilities as a result of the pandemic.

Family caregiving falls heavily on some women of color too. Almost one in five Black women (18%) report caring for someone who needed special assistance prior to the pandemic, significantly higher than the 12% of White women. Eighteen percent of Hispanic women say they have had to take on new caregiving responsibilities since the pandemic started. Nearly one in ten Hispanic women workers (9%) say they have had to take time off work because they were caring for a family member quarantining from or sick with COVID-19.

For some caregivers, the pandemic compounded existing responsibilities. Among women and men who were already caring for someone with special needs prior to March 1, 2020, many said they gained additional caregiving responsibilities as a result of the pandemic (44% and 37% respectively).

Most mothers report the pandemic has taken a toll on their mental health. A majority of mothers of children under 18 (54%) say that worry or stress related to coronavirus has affected their mental health, significantly higher than men with children under 18 (35%) (Table 3). Three in four mothers characterize the impact as major or moderate, as do 55% of fathers. More than half of mothers who are employed (56%) say that pandemic-related stress and worry have affected their mental health, as do 50% of those who are not working. However, only 16% of mothers with children under 18 have sought mental health care, with a significant difference between those who are single (21%) and partnered (14%).

Conclusion

The COVID-19 pandemic has had a disproportionate impact on women in a number of ways. While many have had to put themselves and their families at risk to work on the frontlines as essential workers, a number of women have also had to take on new homeschooling and caregiving responsibilities that have placed additional burdens on their work or resulted in their leaving the workforce entirely. Disparities that existed prior to the pandemic have been magnified and a larger share of women of color and those who are low-income have had to quit their jobs, as well as take on additional caregiving responsibilities. As the nation begins to turn a corner on the pandemic, more schools are re-opening but many remain closed or with limited hours, still placing additional childcare burdens on parents and limiting the ability of many mothers to work at the same level as pre-pandemic. Furthermore, as more businesses reopen, women’s roles in the workplace will not necessarily be restored at the same pace. Some still have additional childcare responsibilities, some may be handling COVID-related illnesses for themselves or family members as well as the major mental health toll on women and their families. Some jobs may not be available anymore, some women had to scale back their workload and lost income, and many women will have missed out on career advancement opportunities. Policies such as paid leave and the child tax credit in the newly enacted federal stimulus bill can support women in caring for their families, obtaining needed health care and balancing work and family responsibilities in this unprecedented time.

The 2020 KFF Women’s Health Survey was designed and analyzed by researchers at the Kaiser Family Foundation (KFF) of a representative sample of 4,805 adults, ages 18-64 years old (3,661 women and 1,144 men). The survey was conducted online and telephone using AmeriSpeak®, the probability-based panel of NORC at the University of Chicago. U.S. households are recruited for participation using address-based sampling methodology and initial invitations for participation are sent by mail, telephone, and in-person interviews. Interviews for this survey were conducted between November 19 and December 17, 2020, among adults living in the United States. KFF paid for all costs associated with the survey.

The sample for this study was stratified by age, race/ethnicity, education, and gender as well as disproportionate stratification aimed at reaching uninsured women, women who identify as LGBT, and women 18-49 years old. The sampling also took into consideration differential survey completion rates by demographic groups so that the set of panel members with a completed interview for a study is a representative sample of the target population. Interviews were conducted in English and Spanish online (4,636) and via the telephone (169).

A series of data quality checks were run and cases determined to be poor-quality, as defined by surveys with a length of interview of less than 33% of the mean length of interview and with high levels of question refusal (>50%) were removed from the final data (n=96). Weighting involved multiple stages. First, the sample was weighted to match estimates for the national population from the 2020 Current Population Survey on age, gender, census division, race/ethnicity, and education. The second round of weights adjusted for the study’s sampling design. All statistical tests of significance account for the effect of weighting.

The margin of sampling error including the design effect for the full sample of women is plus or minus 2 percentage points. Numbers of respondents and margins of sampling error for key subgroups are shown in the table below. For results based on other subgroups, the margin of sampling error may be higher. Sample sizes and margins of sampling error for other subgroups are available by request. Note that sampling error is only one of many potential sources of error in this or any other public opinion poll.

1. People of Hispanic origin may be of any race but are categorized as Hispanic throughout this brief, while other groups are non-Hispanic. ↩︎
2. The federal poverty level (FPL) in 2020 was $26,200 for a family ofu00a0four.u00a0 ↩︎

Key Takeaways

• Women are more likely to have gone without health care during the pandemic compared to men, and women with health and economic challenges prior to the pandemic have experienced worsening health conditions as a result of skipping health care services during the pandemic. These gaps in care could translate into higher numbers of women experiencing severe health issues after the health emergency from the pandemic resolves.
• People have been largely satisfied with their telemedicine interactions.  The share of men and women who at the time of the survey reported having a telemedicine visit during the pandemic tripled since before the pandemic. But not all have had the opportunity or access to telemedicine. Older women, women with higher educational attainment, insured women, and women living in states that have expanded Medicaid are more likely to have had a telehealth visit during the pandemic. While telemedicine played an important role during the pandemic, is it still not clear whether it will continue to be an approach that providers and health consumers will continue to utilize at high rates as the risk of exposure to the coronavirus recedes.
• Women with private insurance and Medicaid are nearly twice as likely to have received a COVID-19 test compared to uninsured women (45% and 41% vs. 28%). Uninsured individuals may have been less likely to seek a test out of fear of cost, which could have implications for uptake of COVID-19 vaccines, even though all COVID-19 vaccines are free regardless of insurance status.
• The pandemic has had a significant effect on people’s mental health, with 51% of women and 34% of men saying that worry or stress related to the pandemic has affected their mental health. Most say that the impact has been moderate or minor, but almost one-fifth (21% women, 17% men) say the toll has taken a major impact on their mental health. The high need for mental health care has highlighted long-standing gaps in the availability and signals that the demand will likely continue as people begin to process the trauma and loss they have experienced over the past year.

Introduction

The COVID-19 pandemic transformed the way people access health care, with care quickly moving to telehealth. However, with fear of contracting the coronavirus and state emergency declarations limiting non-essential and elective services in the early days of the pandemic, many people have gone without health care services this year. The pandemic has highlighted and aggravated long-standing inequities in healthcare availability and access. This brief provides new data from the KFF Women’s Health Survey, a nationally representative survey of 3,661 women and 1,144 men ages 18-64 (Methodology) that was conducted November 19, 2020 – December 17, 2020. Among several topics related to women’s health and well-being, we asked respondents about experiences during the COVID-19 pandemic. In this brief, we document how experiences accessing health care during the COVID-19 pandemic have differed by gender, age, race/ethnicity, insurance coverage, and income and what this could mean moving forward.

Accessing Health Care During the COVID-19 Pandemic

A higher share of women than men have skipped recommended preventive services in response to the pandemic. When asked about their experiences accessing health care services during the pandemic, a larger share of women than men say they have skipped preventive health services, such as a yearly check-up or routine test (38% vs. 26%) or skipped a recommended medical test or treatment (23% vs. 15%). Nearly half of women who report being in fair or poor health report skipping preventive care (46%) and nearly one-third have skipped recommended tests or treatment (32%) compared to women who reported being in good, very good, or excellent health (36% and 21%, respectively). Those in poor health may view themselves as more at risk for COVID exposure and opt to skip routine preventive care. This gap in care among those with the greatest health problems could portend an increase of the share of patients experiencing more severe health conditions resulting from care that was foregone or delayed during the pandemic.

Surprisingly, women with incomes ≥200% of the federal poverty level (FPL) are more likely to have skipped preventive health services during the pandemic compared to women with incomes <200% FPL (40% vs. 33%, respectively),¹  as well as women with private insurance (39%) and Medicaid (38%) compared to uninsured women (30%). A larger share of women with private insurance (23%) and Medicaid (26%) also report skipping a recommended medical test or treatment compared to uninsured women (18%). In pre-pandemic times, the reverse was true, higher shares of women with lower incomes and who lacked insurance coverage reported skipping care. Women with more resources may be choosing to skip care because of concerns of COVID exposure.

Nearly one in five women in fair or poor health (18%) say they have either not filled a prescription, cut pills in half or skipped doses of medicine because of the COVID-19 pandemic, over twice the percentage of women in good, very good, or excellent health (8%). A higher percentage of low-income women < 200% FPL (14%), uninsured individuals (12%), and those with Medicaid (12%) also report not filling a prescription, cutting pills in half or skipping doses of medicine compared to higher income women (≥ 200% FPL) (8%) and individuals with private insurance (8%).ⁱ

More women than men report not being able to get an appointment because of the pandemic (30% vs. 20%), which could be a result of women being more likely to seek care. A larger share of women in poor or fair health (40%) compared to women in good or excellent health (29%) and low-income women (<200% FPL) (33%) compared to higher income women (≥ 200% FPL) (29%) say they were unable to get appointments. State emergency declarations limited services deemed non-essential or elective and reduced clinic hours and closures have made it particularly difficult to access care during the pandemic.

The pandemic has taken a disproportionate toll on some communities of color. Hispanic women, who have the highest rates of uninsurance, report higher rates of access barriers. Four in ten (40%) say they skipped preventive health services, 36% could not get a medical appointment, and 13% did not fill or skipped doses of prescription medicines because of the pandemic (Figure 2). Targeted outreach to this community regarding the COVID-19 vaccine will be important to ensure they have adequate access to and information about the vaccine, especially that the vaccine is free even without insurance.

Nearly one in 10 women ages 18-25 (8%) and 7% of women ages 26-35 say they delayed or were not able to get birth control due to the COVID-19 pandemic, which is significantly higher than women ages 36-49 (3%) (Figure 3). Women without a high school diploma were more likely to report delaying or not being able to get birth control compared to women with a Bachelor’s degree or higher (10% vs. 5%). Individuals who face delays in contraceptive care could face negative health consequences, such as sexually transmitted infections or unwanted pregnancies. While many clinics developed protocols to distribute contraception to their clients by reducing the need for in-person visits through telemedicine or other distribution approaches, addressing access for younger women and those with few resources will continue to be important.

Delays in accessing health care can translate into worsening health conditions, especially for women with underlying health problems that have been exacerbated by lack of care and those who face economic challenges that have gone without care due to cost. Among those who have skipped receiving health care services, women in fair or poor health, those with Medicaid coverage, and those with low incomes are more likely than other groups of women to report that their condition has worsened as a result of skipping care (Figure 4). Providers should be prepared to respond to greater health needs among their patients who have skipped care.

Telehealth

The rapid expansion of telehealth visits has broadened the ways in which people have been able to access care throughout the pandemic. Our survey defines telehealth/telemedicine as an appointment with a provider conducted by telephone or video conference in place of an in-person visit. For this survey, telemedicine or telehealth does not include email, messages sent through a website, or general online health information. Please see the survey topline for full question wording.

Prior to the pandemic, only about 1 in 10 men (11%) and women (13%) had received care via a telemedicine or telehealth visit. However, as of November/December when the survey was in the field approximately 1 in 3 men (32%) and women (38%) reported having a telemedicine or telehealth visit since March 1, 2020. Women over the age of 25, women with a college degree, insured women, and women living in states that have expanded Medicaid are more likely to have had a telehealth visit during the pandemic (Table 1).

Surprisingly, COVID-related care was not a leading reason for seeking health care through telehealth. Just 8% of women and 9% of men say their most recent telehealth appointment was for COVID-related symptoms. More commonly, women report their most recent telehealth visits were to address a minor illness or injury (21%), management of a chronic condition (18%), and 17% for mental health services. While similar shares of women and men report these reasons for telehealth visits, one notable difference is that one-quarter of men (26%) say their most recent telehealth visit was for a checkup, compared to 12% of women (Figure 5).

When asked to rate the care received at their most recent telemedicine or telehealth visit, most women showed high levels of satisfaction. A larger share of women rated the care they received at their most recent telemedicine or telehealth visit as excellent compared to men (35% vs. 29%). Telehealth care rated as excellent, very good, or good was more common among insured women with incomes ≥ 200% FPL compared to uninsured women with incomes < 200% FPL (91% vs. 86%) and women reporting being in excellent, very good, or good health compared to women who said their health was fair or poor (91% vs. 85%). There were no differences in ratings of telehealth quality by insurance status. Among women, telehealth mental health services received the largest share of excellent ratings and annual check-ups or well woman visits have the smallest share of excellent ratings, although ratings were still very positive with 89% of women rating their most recent telehealth annual check-up or well woman visit excellent, very good, or good (Figure 6). These data suggest most women are very satisfied with the telehealth services they have received and telehealth could continue to be a trusted approach to obtaining some types of health care services for individuals who would prefer or find it easier to access care from their home rather than traveling to a clinic or provider. However, there may be pressure from payers as the pandemic eases to once again limit telehealth due to cost concerns.

In recent years, there has been a proliferation of online companies that offer consumers access to primary care services, including sexual and reproductive health services like sexually transmitted infection testing, as well as prescriptions for everything from contraception to hair loss treatment. While these companies predate the pandemic, most are still relatively new. A small fraction of women (4%) ages 18-49 say they have ordered birth control from a website or app, like Nurx, The Pill Club, Pandia Health, or Planned Parenthood Direct. Younger women ages 18-35 (6%) in particular, are more likely to have ordered contraception from these companies compared to those ages 36-49 (1%). Some of these platforms do not accept insurance or Medicaid and may appeal to some uninsured or low-income women because of their costs. A larger share of Hispanic women (5%) compared to White women (3%) and low-income women (<200% FPL) (6%) compared to higher income women (≥ 200% FPL) (3%) have also used these online contraception platforms (Figure 7). While there is still a relatively small share of women using these online companies to obtain contraception, many of these companies have seen significant growth during the pandemic.

COVID-19 Testing

At the time that we fielded this survey in November-December 2020, four in ten women (42%) ages 18-64 had ever been tested for COVID-19 (Figure 8). Larger shares of younger women ages 18-49 (44%) had been tested compared to older women ages 50-64 (38%). Black (46%) and Hispanic (49%) women are also more likely to have been tested for COVID-19 compared to White women (39%).²  Women with private insurance and Medicaid have been nearly twice as likely to have received a COVID-19 test compared to uninsured women (45% and 41% vs. 28%), which could, in part, reflect uncertainty about or inability to afford out-of-pocket costs of COVID-19 tests without insurance. Women who reported having a regular provider are also more likely to have had a COVID-19 test compared to women who did not who do not have a regular health care provider (44% vs. 33%). Access to regular care results in more timely care and could also have implications for who is able to access COVID-19 vaccines, especially if documentation of a pre-existing condition is necessary for earlier vaccine groups.

A small percentage of women reported trying to get a COVID-19 test and being unable to get one (7%) (Figure 9). While a higher share of younger women report being tested for COVID-19, they are also more likely to report not being able to get a test (8%) compared to women ages 50-64 (5%), which may reflect that younger women are more likely to be seeking tests than older women. A larger share of Asian/Pacific Islander (11%) and Hispanic (9%) women say they were unable to get a COVID-19 test compared to White (6%) women, while the share of Black women (7%) who couldn’t get a COVID test is similar to White women. A higher percentage of low-income women (< 200% FPL) (9%) report being unable to get a COVID-19 test compared to higher income (≥ 200% FPL) women (6%).

Impact on Mental Health

Numerous aspects of the pandemic have caused worry and stress, including concern about exposure to an infected individual, not being able to get tested, losing loved ones and jobs, working from home, and taking on additional childcare and homeschooling responsibilities. Half (51%) of women and about one-third of men (34%) say that worry or stress related to the pandemic has affected their mental health. While most (79% women, 83% men) say that the impact has been moderate or minor, almost one-fifth (21% women, 17% men) say the toll has had a major impact on their mental health (Table 2). Women who are already in fair or poor health are more likely than women in better health to say that the pandemic has had a major impact on their mental health.

Despite the toll of the pandemic on women’s mental health, relatively few (15%) say they sought mental health care. While White women (54%) are most likely to say that stress from the pandemic has affected their mental health, Hispanic (19%) and Asian (19%) women are most likely to report seeking mental health care. Uninsured women (10%) are less likely to seek mental health care compared to women with Medicaid (20%) and private insurance (15%). There is likely to be a growing need for mental health care as people process the trauma and loss they have experienced over the past year.

Conclusion

The COVID-19 pandemic changed how people access and use health care, and while the availability of  options for telehealth care have grown rapidly and most people are satisfied with that care, it remains to be seen how acceptable these services will be as in person care resumes, and the requirements for payment parity are ended as the emergency measures are lifted by state and private sector payors. Many individuals are also having difficulties accessing care in this new environment or choosing not to seek care because of concerns about COVID exposure or risk. The survey finds that women have been more likely than men to forgo health care services during the pandemic, with many reporting worsening health conditions as a result. The expansion of telemedicine and virtual reproductive health care platforms have helped fill some of these gaps, but their reach is still limited, particularly for those without insurance or ready access to the internet or Wi-Fi. The burdens of the pandemic have had a negative impact on women’s mental health, though relatively few have reported seeking mental health care. Policies that promote stable insurance coverage, access to telehealth and mental health services, and the availability of safe in-person care can support women in obtaining needed health care for the remainder of the pandemic and beyond.

The 2020 KFF Women’s Health Survey was designed and analyzed by researchers at the Kaiser Family Foundation (KFF) of a representative sample of 4,805 adults, ages 18-64 years old (3,661 women and 1,144 men). The survey was conducted online and telephone using AmeriSpeak®, the probability-based panel of NORC at the University of Chicago. U.S. households are recruited for participation using address-based sampling methodology and initial invitations for participation are sent by mail, telephone, and in-person interviews. Interviews for this survey were conducted between November 19 and December 17, 2020, among adults living in the United States. KFF paid for all costs associated with the survey.

The sample for this study was stratified by age, race/ethnicity, education, and gender as well as disproportionate stratification aimed at reaching uninsured women, women who identify as LGBT, and women 18-49 years old. The sampling also took into consideration differential survey completion rates by demographic groups so that the set of panel members with a completed interview for a study is a representative sample of the target population. Interviews were conducted in English and Spanish online (4,636) and via the telephone (169).

A series of data quality checks were run and cases determined to be poor-quality, as defined by surveys with a length of interview of less than 33% of the mean length of interview and with high levels of question refusal (>50%) were removed from the final data (n=96). Weighting involved multiple stages. First, the sample was weighted to match estimates for the national population from the 2020 Current Population Survey on age, gender, census division, race/ethnicity, and education. The second round of weights adjusted for the study’s sampling design. All statistical tests of significance account for the effect of weighting.

The margin of sampling error including the design effect for the full sample of women is plus or minus 2 percentage points. Numbers of respondents and margins of sampling error for key subgroups are shown in the table below. For results based on other subgroups, the margin of sampling error may be higher. Sample sizes and margins of sampling error for other subgroups are available by request. Note that sampling error is only one of many potential sources of error in this or any other public opinion poll.

1. The federal poverty level (FPL) in 2020 was $26,200 for a family of four.u00a0 ↩︎
2. People of Hispanic origin may be of any race but are categorized as Hispanic throughout this brief, while other groups are non-Hispanic. ↩︎